RESUMO
BACKGROUND: Recent advances in computational pathology have shown potential in predicting biomarkers from haematoxylin and eosin (H&E) whole-slide images (WSI). However, predicting the outcome directly from WSIs remains a substantial challenge. In this study, we aimed to investigate how gene expression, predicted from WSIs, could be used to evaluate overall survival (OS) in patients with lung adenocarcinoma (LUAD). METHODS: Differentially expressed genes (DEGs) were identified from The Cancer Genome Atlas (TCGA)-LUAD cohort. Cox regression analysis was performed on DEGs to identify the gene prognostics of OS. Attention-based multiple instance learning (AMIL) models were trained to predict the expression of identified prognostic genes from WSIs using the TCGA-LUAD dataset. Models were externally validated in the Clinical Proteomic Tumour Analysis Consortium (CPTAC)-LUAD dataset. The prognostic value of predicted gene expression values was then compared to the true gene expression measurements. RESULTS: The expression of 239 prognostic genes could be predicted in TCGA-LUAD with cross-validated Pearson's R > 0.4. Predicted gene expression demonstrated prognostic performance, attaining a cross-validated concordance index of up to 0.615 in TCGA-LUAD through Cox regression. In total, 36 genes had predicted expression in the external validation cohort that was prognostic of OS. CONCLUSIONS: Gene expression predicted from WSIs is an effective method of evaluating OS in patients with LUAD. These results may open up new avenues of cost- and time-efficient prognosis assessment in LUAD treatment.
RESUMO
The move toward early detection and treatment of cancer presents challenges for value assessment using traditional endpoints. Current cancer management rarely considers the full economic and societal benefits of therapies. Our study used a modified Delphi process to develop principles for defining and assessing value of cancer therapies that aligns with the current trajectory of oncology research and reflects broader notions of value. 24 experts participated in consensus-building activities across 5 months (16 took part in structured interactions, including a survey, plenary sessions, interviews, and off-line discussions, while 8 participated in interviews). Discussion focused on: 1) which oncology-relevant endpoints should be used for assessing treatments for early-stage cancer and access decisions for early-stage treatments, and 2) the importance of additional value components and how these can be integrated in value assessments. The expert group reached consensus on 4 principles in relation to the first area (consider oncology-relevant endpoints other than overall survival; build evidence for endpoints that provide earlier indication of efficacy; develop evidence for the next generation of predictive measures; use managed entry agreements supported by ongoing evidence collection to address decision-maker evidence needs) and 3 principles in relation to the second (routinely use patient reported outcomes in value assessments; assess broad economic impact of new medicines; consider other value aspects of relevance to patients and society).